系统仿真学报 ›› 2021, Vol. 33 ›› Issue (2): 306-314.doi: 10.16182/j.issn1004731x.joss.19-0226

• 仿真建模理论与方法 • 上一篇    下一篇

一种改进的萤火虫算法及在洗出优化中的应用

王辉, 吕兴顺*   

  1. 中国民航大学 航空工程学院,天津 300300
  • 收稿日期:2019-05-24 修回日期:2019-10-15 出版日期:2021-02-18 发布日期:2021-02-20
  • 通讯作者: 吕兴顺(1993-),男,硕士,研究方向为飞行仿真技术和智能算法与控制。Email: mike_simon2000@163.com
  • 作者简介:王辉(1966-),男,教授,博士,研究方向为飞行仿真技术和流体传动及控制。Email: 1294127743@qq.com
  • 基金资助:
    国家自然科学基金委员会与中国民用航空局联合资助项目(U1733128)

An Improved Firefly Algorithm and its Application in Washout Optimization

Wang Hui, Lü Xingshun*   

  1. College of Aeronautical Engineering, Civil Aviation University of China, Tianjin 300300, China
  • Received:2019-05-24 Revised:2019-10-15 Online:2021-02-18 Published:2021-02-20

摘要: 为提高萤火虫算法精度,解决该算法迭代步长固定易陷入局部最优等问题,提出一种改进的萤火虫算法—极值优化萤火虫算法(Extremal Optimization Firefly Algorithm,EOFA)。EOFA是将极值动力学算法强大的局部搜索能力与萤火虫算法的强搜索性相结合,采用倒S型函数的迭代步长,提高萤火虫算法的寻优能力。函数寻优测试的仿真结果表明:改进的EOFA相较于萤火虫算法以及粒子群算法都具有更好的寻优性能。将该改进算法应用洗出算法参数优化中,得到了满意的效果。

关键词: 萤火虫算法, 极值动力学算法, 洗出算法, 感知误差, 参数优化

Abstract: To improve the accuracy of the firefly algorithm (FA) and solve the problem of fixed iteration step of the algorithm and easy to fall into local optimum, an improved firefly algorithm is proposed, i.e., EOFA. The EOFA algorithm combines the strong local search ability of extremal optimization algorithm with the strong search ability of firefly algorithm, and adopts the iterative step size of inverted s-type function to improve the optimization ability of the firefly algorithm. The simulation results of function optimization test shows that the improved EOFA algorithm has better optimization performance than firefly algorithm and particle swarm optimization (PSO) algorithm. The improved algorithm is applied to parameter optimization of washout algorithm, and satisfactory result is obtained.

Key words: firefly algorithm, extremum optimization, wash out algorithm, perception error, parameter optimization

中图分类号: